On 6/30/22 3:12 AM, Dylan Yudaken wrote: > This series adds support for multishot recv/recvmsg to io_uring. > > The idea is that generally socket applications will be continually > enqueuing a new recv() when the previous one completes. This can be > improved on by allowing the application to queue a multishot receive, > which will post completions as and when data is available. It uses the > provided buffers feature to receive new data into a pool provided by > the application. > > This is more performant in a few ways: > * Subsequent receives are queued up straight away without requiring the > application to finish a processing loop. > * If there are more data in the socket (sat the provided buffer > size is smaller than the socket buffer) then the data is immediately > returned, improving batching. > * Poll is only armed once and reused, saving CPU cycles > > Running a small network benchmark [1] shows improved QPS of ~6-8% over > a range of loads. I have applied this, changing ->addr2 to ->ioprio for the flags bit as per the io_uring-5.19 branch. Pretty excited about recv multishot. I think it's an elegant model, and it has really nice performance improvements as well! -- Jens Axboe
